Theriogenology 83 (2015) 616–624

Contents lists available at ScienceDirect

Theriogenology

journal homepage: www.theriojournal.com

Characterization of the estrous cycle of Asinina de Miranda jennies (Equus asinus)

M. Quaresma a,b,c,*, R. Payan-Carreira c,d a Veterinary Teaching Hospital, Universidade de Trás-os-Montes e Alto Douro (UTAD), Vila Real, Portugal b Associação Para o Estudo e Proteção do Gado Asinino (AEPGA), Atenor, Portugal c Centro de Estudos Ciências Agrárias e Veterinárias (CECAV), Universidade de Trás-os-Montes e Alto Douro, Vila Real, Portugal d Zootechnics Departement, Universidade de Trás-os-Montes e Alto Douro (UTAD), Vila Real, Portugal article info abstract

Article history: This study aims to characterize the estrous cycle of Asinina de Miranda jennies in the Received 30 May 2014 breeding season, on the basis of data collected from serial ultrasonographic examination Received in revised form 17 October 2014 and serum progesterone determinations in 14 females during a total of 33 cycles. The Accepted 27 October 2014 length of the interovulatory interval was 23.8 0.55 days, the diestrus and estrus lasting 17.9 0.46 days and 6.65 0.30 days, respectively. Age and body condition score (BCS) Keywords: affected the length of the interovulatory intervals; BCS also influenced the diestrus length Donkey and the time in heat after ovulation (P > 0.05). The incidence of single, double, and triple Reproduction Body condition score ovulations was 57.58%, 36.36%, and 6.06%, respectively. Multiple ovulations affected > Ovulation neither the length of the interovulatory interval nor the individual cycle stages (P 0.05) Follicle but lengthened the interval from the beginning of estrus to the last ovulation (P ¼ 0.01). When combined with age, higher BCS affected the ovulation rate (P ¼ 0.001). Deviation of the dominant follicle occurred around Day 8.7 (Day 0 ¼ ovulation) when both single and multiple ovulations were considered. The dominant follicle was larger at divergence in single ovulators (19.18 0.97 mm) compared with that in multiple ovulators (18.05 1.16 mm). The overall maximum follicular diameter before ovulation was smaller in multiple ovulatory cycles than that in single ovulatory cycles (37.2 0.83 mm vs. 40.2 1.41 mm, respectively; P ¼ 0.03). The daily growth rate of dominant follicles was independent of the ovulation rate (P > 0.05) for the intervals before and after the estrus onset. The dominant follicle size and the follicle growth rates were independent of BCS (P > 0.05). Data collected in this study revealed resemblances between Mirandese and other Iberian and Brazilian breeds with regard to estrous cycle characteristics. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction leisure activities such as asinotherapy. With around 500 females available for reproduction, the breed exhibits very The Asinina de Miranda is an endangered breed of low rates of reproduction. Less than a quarter of existing donkey originating from the far northeast of Portugal. It is Asinina de Miranda jennies have had foals registered in the characterized by having a long bay coat, a height of greater stud book [1]; the average foaling rate over the past than 130 cm, and a calm temperament, which makes it 10 years has been close to 50 foals a year, although this especially suited to agricultural work, milk production, and number has increased to around 70 foals a year in the past 2 years [2]. Average age at the first foaling has increased in recent decades, and for a large proportion of females, * Corresponding author. Tel.: þ351 962 615 727; fax: þ351 259 350 663. introduction into reproduction has been postponed until a E-mail address: [email protected] (M. Quaresma). later age, when fertility tends to decline and reproductive

0093-691X/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2014.10.021 M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 617 disorders become more prevalent [3]. The fact that most estrus detection up to ovulation; the laterality of ovulation; traditional owners show little interest in breeding seems to and the length of time in heat after ovulation. Furthermore, be the reason for this late entry into reproduction, together the putative influences of endogenous factors such as age with the lack of any need to replace working animals [1,2]. and body condition score (BCS) were also tested. More recently, under the guidance of the National Breeding Association (Associação Para o Estudo e Proteção do Gado 2. Materials and methods Asinino) and following new trends such as breeding fe- males for milk production, a growing number of young 2.1. Animals, management, and sample collection jennies are now being put into breeding [1,2]. The Asinina de Miranda is an endangered breed, despite This study used 14 nonpregnant clinically healthy attempts to raise interest in it by identifying alternative jennies of the Portuguese Asinina de Miranda breed of uses for these animals. It has been suggested that the donkey. The jennies were aged from 3 to 18 years: six number of females breeding each year be increased and young jennies aged between 3 and 5 years, six adult fe- that jennies should enter into reproduction earlier than males aged between 6 and 8 years, and two older females currently usual in an effort to increase foaling [1,2]. As there aged greater than 15 years. This distribution attempts to is such a limited amount of available information, this reflect the age pyramid seen in breeding females [2];the confirms the need for greater research into the reproduc- small number of females in the older group was due to tive physiology of the species. Milk production is one po- exclusion because of ovarian diseases. All the jennies had tential use of donkeys that has recently been exploited and BCS ranging from four to seven on a nine-point scale (5.7 could provide the means of preventing extinction for many points on average). The body condition score was regu- breeds, including the Asinina de Miranda; however, pro- larly evaluated during the breeding season, at 5-week ductive optimization requires greater knowledge of the intervals [18]. donkey’s reproductive physiology to increase foaling rates Data were collected during two breeding seasons, from [4]. Detailed knowledge of the characteristics of the estrous April to the late September, using a group of seven different cycle in Asinina de Miranda jennies is vital to improve the animals each year. The regular estrous cycles for both the reproductive management of the breed and increase the groups were studied from April to June, producing a total of reproductive efficiency. Furthermore, knowledge on how 33 estrous cycles. Most jennies were found to have two follicles develop and on their size at specific moments in successive estrous cycles; only five females, two from the this process (such as at deviation or ovulation) is funda- first year’s group and three from the second, recorded three mental to pharmacologic manipulation of the cycle or cycles. Complete previous reproductive histories were exogenous induction of ovulation. generally unknown in respect to previous pregnancies, but Jennies are similar to mares in many reproductive as- none of the females had foaled in the preceding breeding pects but tend to have longer breeding seasons [5,6] and season. The existence of regular estrous cycles was longer diestrus phases [7]. Consequently, donkeys have confirmed before the onset of the study. All the jennies longer interestrous intervals [5,8,9], similar to those re- were considered potentially fertile after a breeding ported for ponies [7]. The estrus length is similar among soundness examination. jennies, ponies, and mares [7,10,11], ovulation usually The animals were housed in Vila Real, Portugal (4117 0N occurring 1 to 2 days before the end of estrous behavior 7440W), in the university facilities, and kept under natural [6,9,10], as is also found in mares [12]. photoperiod. All the animals were routinely vaccinated for There are a few studies available on the characteristics equine influenza and tetanus (ProteqFlu-Te; Merial S.A.S., of the estrous cycle for other European donkey breeds Lyon, France) and dewormed every 6 months with 200-mg [7,11,13,14]. However, there have only been a limited ivermectin (Noromectin Oral Paste; Norbrook Laboratories, number of comparative studies on differences in donkey Northamptonshire, UK) per kg of body weight. The jennies breeds, like those known to exist in breeds of horse [15], were kept in a 2500-m2 paddock, with a 50-m2 shelter surveying characteristics such as the rate of ovulation and offering year-round protection from rain, sun, and wind. the prevalence of multiple ovulations, the size of the The animals were fed according to accepted protocols [19], ovulatory follicle, and the moment of ovulation within the consisting of 5 to 7 kg of hay and straw per jenny twice follicular stage. Such characteristics are yet to be deter- daily, which corresponded to a dry matter intake of be- mined in Asinina de Miranda. To implement conservation tween 1.5% and 2% of body weight, supplemented with 1 kg programs aimed at rescuing the breed, it is crucial that of concentrate, divided into two daily portions. Clean fresh assisted reproduction is considered, with particular regard water was always available. The animals were handled in to follicular development and the putative factors that may accordance with EU Directive 2010/63/EU for animal affect it [16,17]. experiments. This study aims to characterize the estrous cycle of The females were group teased daily by a male with a Asinina de Miranda jennies during the breeding season, good libido, and their estrous behavior was classified as including the lengths of the interovulatory interval and of follows [7,20]: The female was considered to be in estrus or the estrus and diestrus stages; the ovulation rate (number receptive if she exhibited mouth clapping together with at of ovulation per estrus) and the prevalence of multiple least one of the following signs during the teasing period: ovulations; the maximum follicular size before ovulation, winking (rhythmic eversion of the vulvar labia with expo- considering both single and multiple ovulations; the sure of the clitoris) and urinating, raising the tail, and pattern of final follicular growth, from the beginning of posturing. In contrast, nonreceptivity behavior included (1) 618 M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 tail down (holding tail down between hind legs when each cycle in the case of multiple ovulations. Sequential mounted), (2) lack of interest (no positive or negative re- US records were used to establish the moment of sponses to the presence or teasing of the jack), and (3) ovulation as the midtime between two US scans when a refusing the jack by moving away or kicking. Clapping alone dominant follicle ceased to be observed during estrus. or combined with kicking or moving was considered to The beginning of estrus was set at the moment when the indicate a transitional stage into or out of estrus but not female first showed signs of receptivity to the male, with recorded as receptive behavior. P4 levels below 1 ng/mL, whereas the end of estrus was For teasing, the male was placed in a paddock adjacent considered to be the moment when the jenny refused the to the jennies, separated by a wire fence. The behavior of jack. Diestrus corresponded to the period when serum P4 the females was observed and recorded for 30 minutes; levels remained above 1 ng/mL and the female refused thereafter, the jack was removed to a box within a closed the jack [22]. The ovulation rate was defined as the building, 400 m away from the females. number of ovulated follicles, on the basis of US obser- For progesterone (P4) measurements, blood samples vation of the collapse of the preovulatory follicle(s) and were collected by venipuncture of the jugular into serum loss of greater than 90% of fluid by the time of the next gel tubes (S-Monovette; Sarstedt, Nümbrecht, Germany), examination [23]. preceding the ultrasound (US) session and placed in ice. Samples were centrifuged after collection at 2500 g for 2.3. Statistical analysis 10 minutes; serum was harvested and stored at 20 C until assayed. Serum P4 concentrations were determined Statistical analyses were performed using the IBM SPSS by chemiluminescent immunoassay (IMMULITE 1000; Statistics 20 software for Windows. The estrous cycles were Siemens Medical Solutions Diagnostics, Los Angeles, CA, normalized to the day of ovulation (Day 0); in the case of USA), using a commercial P4 kit (Siemens IMMULITE Pro- multiple ovulations, Day 0 was set at the day of the last gesterone Kit) and commercially available reagents (all ovulation. For graphical representations, the normalized from Siemens Healthcare Diagnostics, Amadora, Portugal). period was defined as the 12 days after ovulation (Day 0). Interassay coefficient of variance for the controls (CON4, Data for the lengths of the interovulatory interval, diestrus, CON5, and CON6 for low, intermediate, and high controls, and estrus; the size of dominant follicles; the time in estrus respectively; Multivalent Control Module; Siemens) ranged after ovulation; and the final follicular growth rate are from 1.3% and 1.5% for the lower and intermediate controls presented as mean standard error. to 4.6% for the high control. To validate the P4 kit for An ANOVA test was conducted, followed by a Bonfer- donkeys, serial dilutions in the buffer of a blood sample roni post hoc test for means comparison, to analyze the obtained from a 40-day pregnant jenny were made. The effect of BCS and age on the length of each cycle stage, the coefficients of regression obtained were 96%. ovulation rate, the growth rate and follicle size, the total level of P4, and the time in heat after ovulation. Total 2.2. Ultrasound assessment of reproductive activity secretion of P4 during diestrus was assessed by estimating the area under curve, applying the trapezoidal rule, i.e., During the trials, the jennies’ estrous activity was calculating the DX (Y1 þ Y2)/2, using Microsoft Excel routinely surveyed every other day in diestrus and at 8- to 2010 for Windows. Furthermore, a covariance analysis 12-hour intervals in estrus by transrectal palpation fol- was used to explore the effect of BCS (main effect) and age lowed by US examination of the genital tracts using a (covariable) on the ovulation rate. Possible correlations linear-array US scanner equipped with a 5-MHz linear between the ovulation rate and the length of the cycle transducer (Shenzhen Veterinary US scanner), according to phases, the follicular growth rate and follicle size, the total the procedures described by Ginther [15]. The scanner was P4 level, and the time in heat after ovulation were connected to a video camera (DCR-HC96E; Sony), and all US analyzed by the Pearson chi-square test. Differences and scans were recorded for subsequent analysis. correlations were regarded as significant at a P value of The diameters of the ovarian dominant follicles were less than 0.05. obtained retrospectively from the average of the nar- Proportional differences were calculated to determine rowest and widest dimensions in selected US scan images, whether the differences in the proportions of multiple considering only the follicular antrum. One single oper- ovulations in individual animals were significant. ator established follicular size measurements, using ImageJ software (http://imagej.nih.gov/ij/index.html)on 3. Results fixed-frame images. A dominant follicle was defined as the one deviating from the other growing follicles, and 3.1. Estrous behavior becoming the largest in the ovary, whether or not it ovulated [21]. The dominant follicle, or follicles in the case During this study, all jennies in estrus showed homo- of multiple ovulations, was considered ovulatory if it typical signs of estrous behavior (i.e., characteristic for the reached ovulation. Day 0 of the cycle was set as the day of species) such as mouth clapping, clitoral winking, ovulation or, in the case of multiple ovulation, as the day posturing, or showing increased interest toward the male. of the last ovulation. Often, the jennies also exhibited heterotypical behavior The interovulatory interval was defined as the interval (i.e., signs of estrous behavior shared among different (in days) between estrus-associated ovulation in succes- species), such as the flehmen response, sniffing, chasing sive cycles or as the period between the last ovulation of other females, or standing to be mounted. Mouth clapping M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 619 was the first suggestive sign of the approach of estrus, and 53 hours after ovulation, for an average period of it was also the last sign to disappear after ovulation. 26.3 3.27 hours. In the other 12 cycles, the interval be- tween ovulation and the end of estrous behavior was either 3.2. Length of the estrous cycle stages shorter than 12 hours (n ¼ 10) or it occurred before the last ovulation was detected, as in the case of a double ovulation The present study surveyed a total of 33 estrous cycles. that displayed an interval of 45.5 hours between ovulations Cases of anovulatory estrus or of split estrus were not or in a triple ovulation recording an interval of 114 hours observed. The length of the interovulatory interval was between the first and the last ovulation. No significant 23.8 0.55 days, ranging from 17.6 to 34.7 days. The differences (P ¼ 0.508) were found for time in heat after the lengths of diestrus and estrus were 17.9 0.46 days (11.6– last ovulation between single (23.7 5.06 hours) and 27 days) and 6.65 0.30 days (3.15–9.71 days), respectively multiple ovulations (26.25 3.27 hours). Although animals (Table 1). No significant effect of age on the lengths of displaying higher BCS tended to cease estrous behavior estrus (P ¼ 0.682) or diestrus (P ¼ 0.101) was found, sooner after ovulation, the BCS did not significantly affect although longer interovulatory intervals (P ¼ 0.032) were the number of hours in heat after ovulation (P ¼ 0.05). reported in older jennies when compared with those in Nevertheless, longer estruses were linked to a lower younger females. Higher BCS led to longer interovulatory number of hours in estrus after ovulation (P ¼ 0.028). intervals (P ¼ 0.022) and diestrus (P ¼ 0.003) but did not affect the length of estrus (P ¼ 0.944). During the period 3.3. Prevalence of multiple ovulations surveyed, no individual variations were observed neither in the length of interovulatory intervals nor in the length of Of the 33 cycles analyzed, 57.58% (n ¼ 19) were single any phases of the cycle. The ovulation rate did not correlate ovulators and 42.42% (n ¼ 14) multiple ovulators, of which with the length of the interovulatory interval (P ¼ 0.990) or 12 (36.36%) were double ovulations and 2 (6.06%) were diestrus (P ¼ 0.169). However, the estrus was longer in triple ovulations. The number of cycles with multiple multiple ovulators than that in single ovulators (5.22 0.40 ovulations was significantly higher (P ¼ 0.02) in four vs. 6.96 0.33, respectively; P ¼ 0.03). In addition, the jennies, together producing 64.3% (9 of 14) of the multiple ovulation rate correlated positively with the period from ovulations recorded in this study. These animals were the beginning of estrus to the last ovulation (P ¼ 0.01). evenly distributed between young and mature groups, and In general, ovulation occurred less than 15 hours before their BCS ranged from 4 to 5.5 points at the moment of the end of the estrus, but jennies maintained estrous multiple ovulations. For these females, the prevalence of behavior for a variable period after ovulation (Table 1). In multiple ovulatory cycles was significantly higher than that 21 of the 33 cycles studied, estrus lasted between 4 and in the other multiple-ovulating jennies (81.8% vs. 40%,

Table 1 Characteristics of the estrous cycle in Asinina de Miranda jennies in the breeding season.

Type of ovulation n Parameter (days) Mean standard error Minimum, maximum Single 19 Length Interovulatory interval 23.8 0.78 19.5, 34.10 Estrus 5.97 0.37a 3.15, 8.89 Diestrus 18.6 0.65 15.0, 27.00 Intervals Onset of estrus to ovulation 5.22 0.40a 3.15, 8.89 Ovulation to the end of estrus 0.74 0.17 0.00, 1.87 Double 12 Length Interovulatory interval 23.8 0.45 20.5, 26.10 Estrus 7.30 0.44a 4.34, 9.71 Diestrus 17.0 0.32 14.2, 18.60 Intervals Onset of estrus to ovulation 6.80 0.27a 5.58, 8.72 Ovulation to the end of estrus 0.496 0.35 2.56, 2.23 Triple 2 Length Interovulatory interval 24.10 6.60 17.60, 30.70 Estrus 7.82 1.50 6.32, 9.32 Diestrus 16.20 4.62 11.60, 20.90 Intervals Onset of estrus to ovulation 7.91 1.94 5.97, 9.85 Ovulation to the end of estrus 0.09 0.44 0.53, 0.35 Overall 33 Length Interovulatory interval 23.80 0.55 17.60, 34.70 Estrus 6.56 0.30 3.15, 9.71 Diestrus 17.90 0.46 11.60, 27.00 Intervals Onset of estrus to ovulation 5.96 0.31 1.70, 9.85 Ovulation to the end of estrus 0.60 0.16 2.56, 2.23

a Differences were considered significant at a P value of less than 0.05 level. 620 M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 respectively). No influences were found for age or BCS in 19.18 0.97 mm, whereas in multiple ovulators, it occurred the prevalence of multiple ovulations in these four animals. on Day 8.92 0.23 before ovulation, regardless of the order In single ovulators, a nonsignificantly higher frequency of of follicle ovulation, for a follicle diameter of ovulations occurred from the right ovary (57.9%; n ¼ 11) 18.05 1.16 mm. The average size of the dominant follicle compared with the left ovary (42.1%; n ¼ 8). In double ovu- at the onset of estrus (Table 2) was 25 0.95 mm, differing lators (n ¼ 12), ovulation occurred from a single ovary on four (P < 0.001) in the case of single and multiple ovulations and three occasions, respectively for the left and right ovaries, (29.20 1.41 mm and 22.40 1.02 mm, respectively); no whereas it occurred from both the ovaries on five occasions. differences were recorded in the average size of the In triple ovulators (n ¼ 2), two of the follicles ovulated from dominant follicle at the onset of estrus between triple the right ovary and the remainder from the left. (23.30 4.24 mm), double (22.20 0.81 mm; P ¼ 0.130), or Of the 14 multiple ovulations (12 double and 2 triple single (29.20 1.41 mm) ovulations (P ¼ 0.456). ovulations), seven were considered synchronous, with an The overall maximum follicular diameter before interval of less than 24 hours between each ovulation. For ovulation was 38.4 0.68 mm (Table 2). It was smaller in the remainder, the mean interval between ovulations was multiple ovulatory cycles (37.20 0.82 mm) than that in 47.7 7.8 hours (41.8 7.85 for double ovulations and single ovulatory cycles (P ¼ 0.03): in single ovulations 59.4 17.9 for triple ovulations). An unbalanced distribu- (n ¼ 19), the mean maximum follicular diameter was tion of ovulations was observed over the length of a day: a 40.20 1.41 mm, contrasting with 36.70 0.86 mm in higher number of ovulations occurred during daytime double ovulations (n ¼ 24) and 38.60 2.39 mm (n ¼ 6) in (63.3%; n ¼ 31) compared with 36.7% (n ¼ 18) of ovulations triple ovulations (Table 2). In the case of multiple ovula- occurred during the night; the distribution was similar tions, the maximum follicular diameter did not vary with between single and multiple ovulations (P ¼ 0.612). Higher the order of ovulation, whether double (P ¼ 0.096) or BCS affected the occurrence of triple ovulations (P < 0.001), triple ovulations (P ¼ 0.942) were considered, though the but generally, it did not affect the occurrence of multiple second follicle to ovulate was usually smaller. In double ovulations (P ¼ 0.410). When combined with age, higher ovulations, maximum follicular diameter was 38.30 1.25 BCS correlated with a higher ovulation rate (P ¼ 0.001). and 35.41 1.08 mm for the first and second ovulatory follicles, respectively, whereas in triple ovulations, it was 3.4. Growth pattern of dominant follicles 38.20 7.8, 37.50 1.38, and 40.00 4.39 mm, respec- tively, for the first, second and third ovulated follicle. Deviation of the dominant follicle occurred close to Day The ovulation rate did not correlate with the daily 9 before ovulation. In single ovulators, deviation occurred growth rate of the dominant follicle neither during the 8.72 0.40 days before ovulation, for a follicle diameter of period from deviation to the onset of estrus (P ¼ 0.854) nor

Table 2 Follicular development pattern in the breeding season for the Asinina de Miranda jennies.

Number of ovulations n Parameter (mm) Mean standard error Minimum–maximum Single 19 Dominant follicle size At deviation 19.18 0.97 13.20–30.32 At onset of estrus 29.20 1.41a 19.30–46.90 MFD at ovulation 40.20 1.05a 31.80–47.90 Daily growth rate From deviation to onset of estrus 2.62 0.15 1.60–4.07 From onset of estrus to ovulation 3.34 0.31 1.66–5.80 Double 24 Dominant follicle size At deviation 17.00 0.95 15.90–25.25 At onset of estrus 22.20 0.81a 15.74–31.79 MFD at ovulation 36.70 0.86a 30.29–44.19 Daily growth rate From deviation to onset of estrus 2.63 1.86 0.05–8.33 From onset of estrus to ovulation 3.10 0.25 1.49–6.02 Triple 6 Dominant follicle size At deviation 21.57 4.62 14.09–43.56 At onset of estrus 23.30 4.24 15.60–43.50 MFD at ovulation 38.60 2.39 30.36–46.03 Daily growth rate From deviation to onset of estrus 2.41 0.48 0.05–3.13 From onset of estrus to ovulation 3.05 0.64 1.57–5.14 Overall 49 Dominant follicle size At deviation 18.46 0.83 15.29–43,50 At onset of estrus 25.00 0.95 15.60–46.90 MFD at ovulation 38.40 0.68 30.29–47.86 Daily growth rate From deviation to onset of estrus 2.60 0.19 0.05–5.83 From onset of estrus to ovulation 3.18 0.18 1.49–5.80

Abbreviations: MFD, maximum follicular diameter. a Differences were considered significant at a P value less than 0.05 level. M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 621 during the period from the beginning of estrus until until Day 15, P4 levels rose and remained above 10 ng/mL. ovulation (P ¼ 0.955). The daily follicular growth rate Progesterone levels start to drop 2 to 3 days before the during the estrus was significantly higher than that in the onset of estrus, around Days 15 and 16 of the cycle. In period from deviation to the onset of estrus in single estrus, P4 levels remained below 0.2 ng/mL until ovulation (P < 0.001), double (P < 0.001), and triple ovulations (Fig. 1). (P ¼ 0.004). Higher follicular size at the onset of estrus and The area under the P4 curve, corresponding to the total higher daily growth rates of the dominant follicle during level of P4 in diestrus, was higher in multiple ovulatory estrus were associated with ovulation of larger follicles cycles than that in single ovulatory cycles (283.5 18.6 vs. (P ¼ 0.001 and P ¼ 0.027, respectively). 272.9 21.5, respectively; 95% confidence interval; Fig. 1; Dominant follicles reached 30 mm in diameter P ¼ 0.001). 4.1 1.13 days before ovulation, in the case of single ovu- lations, or 2.9 2.47 days before ovulation in the case of 4. Discussion multiple ovulations. The follicular growth rate reported a slowdown of 0.124 0.13 mm/h, as estimated by the Donkeys are often described as displaying longer difference in diameter between the last two measurements estrous cycles than horses but similar in length to pony before ovulation; this slowdown did not differ significantly mares [7,13]. This also applies to Asinina de Miranda jennies. between single or multiple ovulations (P ¼ 0.146; n ¼ 23). The present study found that, in spring, the interovulatory Body condition score did not affect the size of the dominant interval for this breed was close to 24 days, which is in follicle at the onset of estrus (P ¼ 0.688) for this group of accordance with similar studies on other breeds. Consid- Asinina de Miranda jennies. Nor did it affect the maximum erable variation for estimates of the estrous cycle in don- follicular diameter before ovulation (P ¼ 0.818) or the daily keys can be found in the available literature, which in part follicle growth rates, during the moments both before might be due to the period surveyed, the age of the jennies, (P ¼ 0.729) and after the onset of estrus (P ¼ 0.564). or the methods used to define the cycle stages. An overall estrous cycle length of 24 to 25 days is 3.5. Serum progesterone currently accepted for the Catalan (24.9 days; [13]), the Anatolian (25 days; [24]), the Brazilian Pêga (24.2 days; [9]) Mean serum P4 levels at 24 hours after ovulation were and Marchador (23 days; [25]), the Mammoth (23.3 days; 0.48 0.14 ng/mL, rising sharply to values of [6]), the Martina Franca (23.6 days; [22]), and the Baudet de 5.56 0.86 ng/mL by postovulatory Day 3. Thereafter, and Poitou (25.8 days; [26]). The present study surveyed estrous cycles mainly during spring (from April to June), but in accordance with studies from other teams, little variation in the length of estrous cycle with season is to be expected in donkeys from spring to autumn [6,9,22]. In this study, the estimated mean estrus length for Asi- nina de Miranda (6.56 0.55 days) was based on basal P4 concentrations combined with the exhibition of typical estrous behavior. This estimate was similar to that reported for other European breeds: 6.7 days for the Martina Franca [22],6.1 2.1 days for the Zamorano-Leones [14], and 5.64 0.2 days for the Catalan [13]. But, it was shorter than that reported for the Baudet de Poitou (7.5 1.2 days) [26] or Brazilian donkeys (7.9 2.5 days) [10]. In jennies, characteristic signs of estrous behavior in the presence of the jack include mouth clapping, posturing, tail raising, urinating, and clitoral winking [20]. The main homotypical signs of estrus detected in our study were similar to those described for jennies in other studies and used to delimit the estrus stage [6,7,13]. The present study was able to obtain a more accurate estimate of the duration of estrus by integrating the behavioral signs of group- teased females with individual P4 measurements, thus overcoming the reported weaknesses of group teasing in donkeys [6]. The mean diestrus length for the Asinina de Miranda was similar to those reported for Mammoth jennies [6] and the Martina Franca [22], but it was slightly shorter than those reported for other breeds: 17.9 0.46 days vs. 19.83 0.36 in the Catalan [13] or 19.3 0.6 for standard Fig. 1. Serum progesterone and dominant follicle growth during the estrous jennies [7]. cycle for the Asinina de Miranda jennies (mean standard error). The black bar corresponds to the length of estrus. Values are presented in separate for In the present study, age did not affect the lengths of single (A) and double ovulations (B). P4, progesterone. estrus and diestrus in Asinina de Miranda jennies. 622 M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624

Nevertheless, older jennies displayed longer interovulatory pregnancies and minimize their risk to the reproductive intervals, in accordance with those reported for mares [12]. efficiency of this breed. This might be associated with slower growth of the The frequency of ovulation from each ovary registered dominant follicle, as argued by Ginther et al. [27]. This in this study was similar for the left and the right ovaries, finding could not be ascertained in the present study in contrast with that previously reported in horses [15] or because of a disproportionate distribution of ages, with a donkeys [10]. Yet Taberner et al. [13] also failed to find predominance of younger jennies. Nonetheless, in the evidence of statistical differences in the frequency of group of females surveyed, the interovulatory interval and ovulation from the left or right ovary in Catalan jennies. the duration of diestrus were significantly affected by BCS: Multiple ovulations may be classified as synchronous, higher BCS lengthened the interovulatory intervals and when ovulations occur at intervals less than 24 hours, or diestrus in Asinina de Miranda jennies. Although changes in asynchronous, if this interval lasts for more than 24 hours. BCS or in metabolites and metabolic hormones such as In Asinina de Miranda jennies, a similar proportion of leptin, insulin, or insulinlike growth factor 1 have been synchronous and asynchronous was observed. This con- associated with follicular activity and mare fertility [28,29], trasts with descriptions of the Catalan breed, most of there is a lack of incontrovertible information available on whose ovulations were asynchronous, with intervals the effect of BCS on conditioned measurements of the ranging from 1 to 9 days [13]. In our study, the maximum duration of each stage of the estrous cycle in cyclic mares interval found between multiple ovulations was 2.48 days, [30,31]. which is longer than that reported for Pêga jennies [9] but Moreover, Fitzgerald and McManus [31] reported resembling that reported for Przewalski’s mares [33] or similar effects of BCS on the characteristics of the estrous standard jennies [5]. cycle, affirming that the length of diestrus and inter- When planning assisted reproductive technologies, ovulatory interval was greater in fat mares (BCS 7) under knowledge regarding development of the dominant folli- controlled management than that in mares with moderate cle, including its size at the onset of estrus, around ovula- BCS. The effect of high BCS on the duration of the estrous tion and its daily growth rate, is fundamental for stages in different studies may incorporate the effect of manipulating the estrous cycle and inducing ovulation. In other parameters, such as the age and breed of the female, Asinina de Miranda, these measurements were similar to the management (controlled vs. free ranging), the physio- those reported in other breeds with which the Portuguese logical status (postpartum, cyclic), the extent of the period breed shares some resemblances in the estrous cycle. considered (the entire year vs. the breeding season), or the In the present study, the dominant follicle was first number of consecutive cycles, limiting the scope of this detected in the ovary as the fastest growing follicle at about discussion. It is possible that this also occurred in the pre- 13 days before ovulation. The mean follicular diameter at sent study, as older mares tend to display higher BCS levels the onset of estrus, corresponding to Days 5 to 6 before than younger mares. ovulation in Asinina de Miranda jennies, was close to Multiple ovulations seem to be higher in donkeys than 25 mm, which is in accordance with that of the Brazilian those in horses [15]; the ovulation rate reported in the Marchador [25] but lower than that reported for the Mar- available literature varies from around 5% to almost 70% tina Franca (around 31.5 mm; [22]) in the same season. [6,7,10,13]. It has been proposed that one main factor for The dominant follicle reached 30 mm around 2.5 to this variation in donkeys might be the breed [13], although 4 days before ovulation, for multiple and single ovulations. no statistical differences were found among three different The average maximum follicular diameter observed in Spanish breeds [14]. In the present study, the prevalence of jennies in the present study was 38.4 mm, which is similar multiple ovulations in Asinina de Miranda jennies was to that reported for the Pêga [9] and Marchador [25] or in 42.42%, of which 36.36% were double ovulations and 6.06% standard jennies [7]. But, it was lower than that recorded in triple. These figures were similar to those reported for the Catalan (close to 45 mm; [13])orMartina Franca jennies three Spanish breeds, the Catalan [13,14], the Andalusian, (43.7 mm; [22]). In contrast to the research carried out by and the Zamorano-Leonês [14] but lower than those for Taberner et al. [13], which fails to provide evidence of a link Mammoth donkeys [6]. with ovulation type (simple vs. multiple), the maximum As previously reported in mares [15] and Catalan jennies follicular diameter was the largest in the cases of single [13], multiple ovulations were highly repetitive in Asinina rather than multiple ovulations. Similar observations have de Miranda females. According to Ginther [15], this suggests also been reported in mares [21] and Brazilian Marchador that it may be a heritable trait. The existence of multiple jennies [25]. ovulations did not affect the interovulatory interval in As expected, the daily follicular growth is higher dur- Asinina de Miranda jennies, although it extended the estrus ing estrus than that in the period between deviation and and the interval from the beginning of the estrus until the onset of estrus, as acknowledged in mares [12].Little ovulation. Our results are supported by comparable de- information is available for donkeys, as most studies have scriptions in Spanish donkey breeds [14]. In the present focused on the follicular growth rate in the 5 days pre- study, the prevalence of multiple ovulations was positively ceding ovulation, which corresponds to estrus. In the affected by BCS as has also been reported in mares [32]. present study, the daily growth rate of the dominant fol- Information gathered on multiple ovulations in Asinina de licle was significantly higher after the onset of estrus Miranda jennies, along with the positive effect of BCS on (3.18 0.18 mm/day) than that in the period before estrus their occurrence, highlights the need to routinely imple- (2.60 0.19 mm/day), independently of the ovulation rate ment an early pregnancy diagnosis service to identify twin considered. Compared with other studies, the mean daily M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624 623 follicular growth during estrus for Asinina de Miranda Acknowledgments jennies was slightly higher than that described for the Brazilian Marchador,(2.39 0.37 mm/day [25])orthe The authors wish to thank Associação Para o Estudo e mare (2.7 mm/day [15]) but lower than that reported in Proteção do Gado Asinino for providing the jennies and Catalan jennies (3.7 mm/day [13]). As previously reported material used in the study and M. Cristina Bastos de Car- for Catalonian donkeys [13], there is a slowdown in the valho for her help in feeding and handling the jennies and daily growth rate of the dominant follicle on the day in collecting data. The authors also thank Dr. Celso Santos preceding ovulation in Asinina de Miranda. Such knowl- for his help with progesterone measurements and Prof. Dr. edge of follicular dynamics is of utmost importance for Luis Ferreira for his help in producing the figure. controlling ovulation in any breed, enabling drug admin- This study was sponsored by the Portuguese Science istration schedules and timing of insemination to be and Technology Foundation (FCT) under projects PEst-OE/ personalized. AGR/UI0772/2011 and PEst-OE/AGR/UI0772/2014. In the present study, the jennies’ BCS affected neither Author contributions: M. Quaresma and R. Payan- the size of the dominant follicle nor its growth pattern. This Carreira conceived the study and participated in its seems to contrast with the work of Gastal et al. [16], which design. M. Quaresma conducted the animal reproductive reported that in mares, the body condition was positively assessment and the sequential blood collection and ultra- linked with the maximum diameter of preovulatory folli- sound examinations. M. Quaresma analyzed and collected cles for the first ovulations of the breeding season. More- data from ultrasound films and interpreted the data. In over, Lemma et al. [17] found that the BCS was positively addition, both the authors were responsible for compiling correlated to the diameter of the preovulatory dominant the literature review, drafting, and finalizing the article. follicle in Ethiopian jennies. The relatively constant mod- Both the authors read and approved the final article. Finally, erate body condition evidenced by the females in the both the authors studied and addressed the issues raised in present study, however, might explain the differences be- the review panel’s comments and together revised the tween our results and those referred to that mentioned article. previously. In general, the cyclic changes in P4 levels in Asinina de Competing interests Miranda resemble those reported in other donkey breeds [9,22,34] and mares [21]. Individual variations are expected None of the authors has any financial or personal rela- both in the onset of P4 peak and in P4 levels, as has also tionship that could inappropriately influence or bias the been described in mares with estrous cycles of similar content of the article. length [35,36]; such variations were associated with dif- ferences in the secretory capacity of the CL and the hor- fi monal catabolic rate and appear to be more signi cant in References the first 5 days of the diestrus [36]. Moreover, the existence of multiple ovulations and their frequency of occurrence [1] Quaresma M, Martins AM, Rodrigues JB, Colaço J, Payan-Carreira R. may also influence the levels of P4 measured. Comparison Pedigree and herd characterization of a donkey breed vulnerable to – of P4 levels in diestrus, using the area under the curve, extinction. Animal 2014;8:354 9. [2] Quaresma M, Martins AMF, Rodrigues JB, Colaço J, Payan-Carreira R. shows that, in our study, they were affected by the number Viability analyses of an endangered donkey breed: the case of the of ovulations, in accordance with data presented by Meira Asinina de Miranda (Equus asinus). Anim Prod Sci 2014;54. (In et al. [9] in Pêga jennies. Press). [3] Quaresma M, Payan-Carreira R, Pires MA, Edwards JF. Bilateral ovulation fossa inclusion cysts in Miranda jennets. J Comp Pathol 4.1. Conclusions 2011;145:367–72. [4] Carluccio A, Contri A, De Amicis I, Robbe D. La riproduzione della fi fattrice asinina: nozione di ginecologia e ostetricia. In: Milonos E, The present study has enabled identi cation of the Polidori P, editors. Latte di asinna produzione, caratteristiche e estrous cycle characteristics of Asinina de Miranda jennies gestione dell’azienda asinina. Brescia, Italy: Fondazione Iniziative during the breeding season. Data collected revealed some Zooprofilattiche e Zootecniche; 2011. p. 91–100. [5] Ginther OJ, Scraba ST, Bergfelt DR. Reproductive seasonality of the resemblances with other Mediterranean and Brazilian jenney. Theriogenology 1987;27:587–92. donkey breeds. It was observed that the BCS was posi- [6] Blanchard TL, Taylor TS, Love CL. Estrous cycle characteristics and tively linked to multiple ovulations and the length of response to estrus synchronization in mammoth asses (Equus asinus americanus). Theriogenology 1999;52:827–34. interovulatory intervals, although the jennies maintained [7] Vandeplassche GM, Wesson JA, Ginther OJ. Behavioral, follicular and a moderate BCS. Furthermore, jennies with higher BCS gonadotropin changes during the estrous cycle in donkeys. Ther- appeared to cease estrous behavior after ovulation faster iogenology 1981;16:239–49. d than those with a lower score. The BCS did not affect [8] Fielding D. Reproductive characteristics of the jenny donkey Equus asinus: a review. Trop Anim Health Prod 1988;20:161–6. estrous and diestrous duration per se and did not seem to [9] Meira C, Ferreira JCP, Papa FO, Tornero MTT, Bicudo SD. Study of the be linked to dominant follicle size and growth rate. The estrous cycle in donkeys (Equus asinus) using ultrasonography and present study also showed that at the onset of estrus, the plasma progesterone concentrations. Biol Reprod Mono 1995;1: 403–10. dominant follicle was about 25 mm in diameter. This [10] Henry M, Figueiredo AEF, Palhares MS, Coryn M. Clinical and study also provides important data on measurements endocrine aspects of the oestrous cycle in donkeys (Equus asinus).J concerning follicular growth for those intending to Reprod Fertil 1987;35:297–303. ’ [11] Carluccio A, Contri A, Mattioli M, Vilani M, Veronesi MC. Proges- manipulate the Asinina de Miranda s cycles for assisted terone and LH plasma concentrations during the estrous cycle in reproduction. Martina Franca donkeys. Reprod Domest Anim 2005;40:P18. 624 M. Quaresma, R. Payan-Carreira / Theriogenology 83 (2015) 616–624

[12] Blanchard TL, Dickson DV, Schumacher J, Love CC, Brinsko SP, [25] Conceição JC, Freitas Neto LM, Aguiar Filho CR, Araújo GHM, Rigby SL. Manual of equine reproduction. Second edition. USA: Oliveira JV, Bartolomeu CC, et al. Ultrasound evaluation of ovarian Mosby; 2003. follicular dynamics in jennies (Equus asinus) during the estrous [13] Taberner E, Medrano A, Peña A, Rigau T, Miró J. Oestrus cycle cycle. Medicina Veterinária 2009;3:7–14. characteristics and prediction of ovulation in Catalonian jennies. [26] Trimeche A, Tainturier D. Étude échographique de la cinétique fol- Theriogenology 2008;70:1489–97. liculaire des baudettes du Poitou pendant l’oestrus au cours du [14] Galisteo J, Perez-Marin CC. Factors affecting gestation length and printemps et de l’été. Rev Med Vet 1995;146:743–8. estrus cycle characteristics in Spanish donkey breeds reared in [27] Ginther OJ, Gastal EL, Gastal MO, Beg MA. Dynamics of the equine southern Spain. Theriogenology 2010;74:443–50. preovulatory follicle and periovulatory hormones: what’s new? J [15] Ginther OJ. Reproductive biology of the mare. Cross Plains, WI: Equine Vet Sci 2008;28:454–60. Equiservices Publishing; 1992. [28] Fradinho MJ, Correia MJ, Grácio V, Bliebernicht M, Farrim A, [16] Gastal MO, Gastal EL, Spinelli V, Ginther OJ. Relationships between Mateus L, et al. Effects of body condition and leptin on the repro- body condition and follicle development in mares. Anim Reprod ductive performance of Lusitano mares on extensive systems. 2004;1:115–21. Theriogenology 2014;81:1214–22. [17] Lemma A, Bekana M, Schwartz HJ, Hildebrandt T. The effect of body [29] Morley S, Murray J. Effects of body condition score on the repro- condition on ovarian activity of free ranging tropical jennies (Equus ductive physiology of the broodmare: a review. J Equine Vet Sci asinus). J Vet Med A Physiol Pathol Clin Med 2006;53:1–4. 2014;34:842–53. [18] Quaresma M, Payan-Carreira R, Silva SR. Relationship between ul- [30] Nagy P, Guillaume D, Daels P. Seasonality in mares. Anim Reprod Sci trasound measurements of body fat reserves and body condition 2000;60-61:245–62. score in female donkeys. Vet J 2013;197:329–34. [31] Fitzgerald BP, McManus CJ. Photoperiodic versus metabolic signals [19] Smith D, Wood S. Donkey nutrition. In: Duncan J, Hadrill D, editors. as determinants of seasonal anestrus in the mare. Biol Reprod 2000; The professional handbook for the donkey. Fourth edition. Ehittet 63:335–40. Books; 2008. p. 1–27. [32] Guillaume D, Salazar-Ortiz J, Martin-Rosset W. Effects of nutri- [20] McDonnell SM. Reproductive behavior of donkeys (Equus asinus). tion level in mares’ ovarian activity and in equines’ puberty. In: Appl Anim Behav Sci 1998;60:277–82. Miraglia N, Martin-Rosset W, editors. Proceedings of European [21] Ginther OJ, Gastal EL, Rodrigues BL, Gastal MO, Beg MA. Follicle Workshop on Equine Nutrition (EWEN 2006, Campobasso, Italy): diameters and hormone concentrations in the development of nutrition and feeding of the broodmare, EAAP publication no. single versus double ovulations in mares. Theriogenology 2008;69: 120. The Netherlands: Wageningen Academic Publishers; 2006. 583–90. p. 315–38. [22] Contri A, Robbe D, Gloria A, De Amicis I, Veronesi MC, Carluccio A. [33] Collins CW, Songsasen NS, Vick MM, Wolfe BA, Weiss RB, Keefer CL, Effect of the season on some aspects of the estrous cycle in Martina et al. Abnormal reproductive patterns in Przewalski’s mares are Franca donkey. Theriogenology 2014;81:657–61. associated with a loss in gene diversity. Biol Reprod 2012;86:1–10. [23] Cuervo-Arango J, Newcombe JR. Ultrasound confirmation of ovula- [34] Carluccio A, Panzani S, Contri A, Tosi U, De Amicis I, Veronesi MC. tion in mares: a normal corpus luteum or a haemorrhagic anovu- Luteal function in jennies following PGF(2alpha) treatment 3 days latory follicle? Reprod Domest Anim 2013;48:105–11. after ovulation. Theriogenology 2008;70:121–5. [24] Kalender H, Aslan S, Schwarzenberger F, Maral NE, Izgur H, [35] Squires EL, Wentworth BC, Ginther OJ. Progesterone concentration Handles J, et al. Uterine involution, follicle development and con- in blood of mares during the estrous cycle, pregnancy and after centrations of plasma progesterone, 20a-OH-progesterone and total hysterectomy. J Anim Sci 1974;39:759–67. estrogen levels during the postpartum period in anatolian donkeys. [36] Satué K, Gardón JC. A review of the estrous cycle and the neuro- Kafkas Univ Vet Fak Derg 2012;18:929–33. endocrine mechanisms in the mare. J Steroids Horm Sci 2013;4:115.